Income Inequality and Political Stability

by Samantha Monk

As escalating civil unrest flares up across the globe, displacing thousands and costing people their lives, attempts to understand the causes of these violent eruptions have also multiplied. In this post, I look for evidence of a correlation between income inequality and political stability.

“Deprivation rationalizations” for these conflicts, revolutions, or uprisings argue that revolutions occur when significant segments of the population are at a strong disadvantage to those in power. Basically, the belief is that disenfranchised groups are motivated to take action against governments that are failing them.

Researchers like Alberto Alesina and Robert MacCulloch have attempted to validate this hypothesis using a variety of variables and statistical techniques. Some have used survey data to evaluate attitudes towards revolt. Others have used political change as a measure of propensity for revolution. Some have created their own variables to try to represent “political stability”, which is a vague concept and thus difficult to quantify. Others have used case studies to highlight examples where economic inequality was not a major factor in regime destabilization; two such articles are summarized here.

Research regarding this issue offers various conclusions. Most of it was done too long ago to factor in of recent events. Revolutions, insurrections, revolts, and widespread protests in Burkina Faso, Ukraine, South Sudan, Hong Kong, and many other spots on the map make it clear that this is still a question worth investigating.

Here, I’ve used data from Polity IV and GINI coefficients to investigate a potential correlation between political instability and income inequality. Polity IV scores country stability based on a variety of factors, including regime type and data about military coups, uprisings, civil liberties, and political rights. A low polity score means the country is highly unstable. This is not a perfect variable; an aggregate variable which includes many factors might insinuate false correlations.

I used GINI coefficient to measure income inequality. A GINI coefficient of zero signifies a perfectly equal distribution of income, while a value of one signifies maximum inequality. There are also critiques of the GINI coefficient. It doesn’t always accurately depict the severity of the inequality. For example:

• If 100 people live in a country, and 50 of them have all the money, then GINI = .5 
• If 75 of them have 25% of the money, and the remaining 25 people have 75%, GINI = .5.

Clearly neither Polity IV nor GINI are perfect measures, but both do a fair job at quantifying these two variables.

I also included log GDP as a control variable, in an attempt to limit the impact that GDP may have in skewing the data to show stable countries as those with lower income inequality. Other important control variables are military expenditure and oil rents. These variables are often related to political instability and income inequality, so it’s worth taking them into account. Some scholars argue that political instability scares leaders into investing in armies and weapons to protect themselves. Other evidence suggests that oil makes the rich get richer (they control the resources initially and then soak up all the profits) while the poor get poorer. Summary statistics for all variables can be found in Table 2 at the end of this post.
Figure 1 shows the effect of Economic Inequality (GINI coefficient) on Political Instability (Polity IV score). Neither the loess line (in black) nor the OLS regression line (in burgundy) fits the data very well, suggesting either a low correlation, or that outliers are dragging the line away from where it really belongs. However, it does look at first glance like lower GINI scores (incomes are fairly equal) do in fact have a slight tendency to be more politically stable.

Table 1 describes my initial multivariate regression analysis. When accounting only for income inequality and political stability, there appears to be a weak but statistically significant correlation between the two variables at issue here. A decline in economic inequality appears to correspond to some degree of decline in political stability. A model which adds oil rents and military expenditures into the mix shows surprisingly similar values. However, when GDP is taken into effect the correlation becomes weaker and is not statistically significant. Instead, we see a strong (and unsurprising) positive correlation between GDP and political stability.

This paper attempted to investigate a potential correlation between economic inequality and political instability. Based on these models, I do not feel comfortable accepting or rejecting the hypothesis that economic inequality and political stability are correlated. Two of the models suggested a slight, possible correlation, while the third emphatically did not.

The most serious limitation to this study is that the measurements used can be interpreted as carrying many different meanings. This research used the Polity IV score for political stability, which has been criticized as subjective and vague, among other things. GINI index, although widely used is also criticized for a variety of reasons, including its sensitivity to population distribution.

Another potential limitation of this analysis is that I picked data from one specific year, 2012. In order to have enough cases for analysis, I had to supplement this with some data from 2010, 2011, and 2013. It may be that these slight departures from accuracy caused some unintended statistical consequences. It may also be more useful to consider data over a period of time, as it is unlikely that the political events of one year alone can accurately describe the long term political stability (or lack thereof) of a government.

In light of current events in many areas of the world, questions about what causes political instability are especially valid, but unfortunately, this research has so far failed to provide any useful hints.

Sources

Alesina, A., and R. Perotti. "Income Distribution, Political Instability, and Investment.” European Economic Review: 1203-228.

Alesina, Alberto, Sule Özler, Nouriel Roubini, and Phillip Swagel. "Political Instability and Economic Growth." Journal of Economic Growth: 189-211.

Campos, Nauro F, and Jeffrey B Nugent. "Who Is Afraid of Political Instability?" Journal of Development Economics: 157-72.

Dutt, Pushan, and Devashish Mitra. "Inequality and the Instability of Polity and Policy." The Economic Journal: 1285-314.

MacCulloch, Robert. "Income Inequality and the Taste for Revolution*." The Journal of Law and Economics: 93-123.

Marshall, Monty G., and Benjamin R. Cole. "State Fragility Index and Matrix 2013." Center for Systemic Peace; see http://www.systemicpeace.org/inscr/SFImatrix2013c.pdf

Midlarsky, M. I. "The Origins of Democracy in Agrarian Society: Land Inequality and Political Rights." Journal of Conflict Resolution, 1992, 454-77.

Muller, E. N., and E. Weede. "Theories of Rebellion: Relative Deprivation and Power Contention." Rationality and Society, 1994, 40-57.

"What Is the Relationship Between Income Inequality and Revolution?" Freakonomics RSS. Accessed March 17, 2014 http://freakonomics.com/2012/09/25/what-is-the-relationship-between-income-inequality-and-revolution/.

World Bank. (2014). Data retrieved November 2014, from World Development Indicators Online (WDI) database.